MLS-C01 Hard Questions — 25 Difficult Questions

Question 1

A team is building a data pipeline to process terabytes of log data daily using Amazon EMR. The data arrives in 5-minute windows and must be available for querying within 30 minutes. The data is originally in gzip-compressed CSV files. Which approach will minimize processing time and cost?

Accepted Answer

Use Amazon EMR with Spark to convert data to Parquet and store in S3, using spot instances for task nodes.. Option B is correct because converting gzip-compressed CSV to Parquet reduces storage size and improves query performance due to columnar storage and predicate pushdown. Using spot instances for task nodes significantly lowers compute cost, while the 30-minute SLA is achievable with Spark on EMR processing 5-minute windows of data.

Answer

Use Amazon EMR with Spark to convert data to Parquet and use on-demand instances.

Answer

Use AWS Glue to convert data to gzip-compressed CSV and query with Athena.

Answer

Use Amazon EMR with Hive to transform data to compressed CSV and store in S3.

Question 2

A company uses Amazon SageMaker to train and deploy machine learning models. The training data is stored in Amazon S3 (Parquet format, 10 TB). The data scientists have been running training jobs using the File mode input, but the jobs are taking too long due to data download time. They want to reduce the training start-up time and overall training time. Which solution is MOST cost-effective and efficient?

Accepted Answer

Configure the SageMaker training job to use Pipe mode, which streams data directly from S3 without downloading to the instance's local storage.. Pipe mode in SageMaker streams training data directly from Amazon S3 to the training algorithm without first downloading it to the instance's local storage. This eliminates the data download step, significantly reducing startup time and overall training time for large datasets like 10 TB. It is the most cost-effective because it avoids the need for larger instances or additional data transfer acceleration services.

Answer

Use S3 Transfer Acceleration to speed up the data transfer from S3 to the training instance.

Answer

Use larger EC2 instances with more vCPUs and memory to speed up the training process.

Answer

Enable Elastic Fabric Adapter (EFA) on the training instances to improve network throughput.

Question 3

A company needs to build a data lake on AWS for analytics. The data includes structured, semi-structured, and unstructured data. The solution must support schema-on-read, provide fine-grained access control, and be cost-effective for storing rarely accessed data. Which THREE services should be used? (Choose THREE)

Accepted Answer

AWS Glue Data Catalog for schema-on-read.. AWS Glue Data Catalog is correct because it provides a centralized metadata repository that enables schema-on-read for data stored in Amazon S3. It allows you to define table schemas and partitions without transforming the underlying data, so analytics tools like Amazon Athena and Amazon EMR can query the data with the schema applied at read time.

Answer

Amazon Redshift for data warehousing.

Answer

Amazon EMR for data processing.

Question 4

A data engineer created an IAM policy to allow a Glue ETL job to read and write objects to an S3 bucket. The ETL job fails when writing data with the error 'Access Denied'. The job is configured to use SSE-S3 (AES256) encryption. What is the likely issue?

Accepted Answer

The policy does not grant s3:PutObject on the bucket itself, which is needed for some write operations.. The error 'Access Denied' when writing to S3 with SSE-S3 encryption typically occurs because the IAM policy lacks the `s3:PutObject` permission on the bucket resource itself. While the policy may grant `s3:PutObject` on the object ARN (`arn:aws:s3:::bucket/*`), some S3 write operations—especially those involving encryption headers or bucket-level checks—also require the permission on the bucket ARN (`arn:aws:s3:::bucket`). Without this, the request is denied even if the object-level permission exists.

Answer

The policy grants s3:PutObject on all buckets, not just the specific one.

Answer

The condition requires objects to be encrypted with SSE-KMS, but the job uses SSE-S3.

Answer

The condition requires objects to use SSE-S3, but the job uses SSE-KMS.

Question 5

A company runs a real-time fraud detection system using Amazon Kinesis Data Streams with 100 shards. Data is consumed by a custom Java application running on Amazon EC2 instances in an Auto Scaling group. The application processes records and writes results to a DynamoDB table. Over the past month, the application has experienced intermittent slowdowns and the DynamoDB write capacity has been fully utilized during peak hours. The team wants to improve throughput without losing the ability to reprocess failed records. The application currently uses the Kinesis Client Library (KCL) with DynamoDB as the lease table. The team is considering the following changes: A. Increase the number of EC2 instances to match the number of shards. B. Switch to using AWS Lambda as the consumer to handle scaling automatically. C. Increase the write capacity of the DynamoDB lease table to handle more workers. D. Use enhanced fan-out to have each consumer receive its own 2 MB/second shard throughput. Which change should the team implement first to address the issue?

Accepted Answer

Use enhanced fan-out to have each consumer receive its own 2 MB/second shard throughput.. The primary bottleneck is DynamoDB write capacity being fully utilized during peak hours. Enhanced fan-out (option B) provides each consumer with a dedicated 2 MB/second read throughput per shard, eliminating the need for consumers to contend for the shared 2 MB/second per shard. This reduces the load on the DynamoDB lease table because workers no longer need to poll for records, which in turn lowers the write operations to the lease table and alleviates the DynamoDB write capacity issue.

Answer

Increase the write capacity of the DynamoDB lease table to handle more workers.

Answer

Switch to using AWS Lambda as the consumer to handle scaling automatically.

Answer

Increase the number of EC2 instances to match the number of shards.

Question 6

A data scientist is performing EDA on a dataset with 1,000 features and 10,000 rows. The target variable is binary. After checking for multicollinearity, the scientist finds many pairs of features with correlation > 0.95. Which action should be taken to prepare the data for modeling?

Accepted Answer

For each highly correlated pair, remove one feature based on domain knowledge or higher correlation with target.. Option C is correct because when features are highly correlated (e.g., > 0.95), they introduce multicollinearity, which can destabilize coefficient estimates in linear models and reduce interpretability. Removing one feature from each correlated pair based on domain knowledge or its correlation with the target variable preserves predictive power while reducing redundancy. This approach is more targeted than PCA, which transforms features into uncorrelated components but sacrifices interpretability and may not align with the binary target.

Answer

Apply PCA to all features to decorrelate them.

Answer

Standardize all features using StandardScaler.

Answer

Randomly drop half of the correlated features.

Question 7

An e-commerce company uses a linear regression model to predict customer lifetime value (LTV). The model shows high variance on the test set, with training RMSE much lower than test RMSE. Which of the following is the MOST effective approach to reduce overfitting?

Accepted Answer

Apply L2 regularization (Ridge regression). High variance (low training RMSE, high test RMSE) indicates overfitting. L2 regularization (Ridge regression) adds a penalty proportional to the square of the coefficients, shrinking them toward zero without eliminating them, which reduces model complexity and improves generalization. This directly addresses overfitting by constraining the model's sensitivity to noise in the training data.

Answer

Use a polynomial kernel in a support vector regressor

Answer

Add more features, including interaction terms

Answer

Increase training data size by duplicating existing samples

Question 8

A data scientist is training a deep learning model using Amazon SageMaker. The training loss is decreasing, but the validation loss starts increasing after 10 epochs. The model is overfitting. Which TWO actions should the data scientist take to reduce overfitting? (Choose 2.)

Accepted Answer

Add dropout layers. Option D is correct because dropout layers randomly deactivate a fraction of neurons during training, which forces the network to learn more robust features and reduces co-adaptation, a common cause of overfitting. This technique is particularly effective in deep learning models trained on SageMaker, where large architectures can quickly memorize training data.

Answer

Increase the number of layers

Answer

Remove L2 regularization

Answer

Increase the number of training steps

Question 9

A data scientist is exploring a dataset with 500 features and 10,000 samples. The data scientist computes the pairwise correlation matrix and finds that many features have correlations above 0.9. The data scientist wants to reduce the dataset to 50 features while preserving as much variance as possible. Which technique should be used?

Accepted Answer

Apply Principal Component Analysis (PCA) and keep the top 50 principal components.. Principal Component Analysis (PCA) is the correct technique because it performs an orthogonal linear transformation that projects the original 500 features into a new coordinate system where the axes (principal components) are ordered by the variance they capture. By keeping the top 50 principal components, the data scientist retains the maximum possible variance in the reduced 50-dimensional space, directly addressing the goal of preserving variance while handling high multicollinearity.

Answer

Remove all but one feature from each group of highly correlated features.

Answer

Use Linear Discriminant Analysis (LDA) to project to 50 dimensions.

Answer

Use t-Distributed Stochastic Neighbor Embedding (t-SNE) to reduce to 50 dimensions.

Question 10

A data scientist is analyzing a dataset of customer reviews. The dataset contains a text column 'review' and a numerical rating from 1 to 5. The data scientist wants to create features for sentiment analysis. Which THREE preprocessing steps should be applied to the text data before feature extraction? (Choose THREE.)

Accepted Answer

Apply stemming to reduce words to their root form.. Option B is correct because stemming reduces words to their root form (e.g., 'running' to 'run'), which consolidates variations of the same word and reduces feature dimensionality. This is a standard preprocessing step before feature extraction in NLP tasks like sentiment analysis, as it helps the model generalize across different word forms.

Answer

Standardize the text data using z-score normalization.

Answer

Tokenize the text into individual words.

Question 11

Which THREE of the following are best practices when performing exploratory data analysis on a dataset with both numerical and categorical features?

Accepted Answer

Check the proportion of missing values for each feature.. Option A is correct because checking the proportion of missing values for each feature is a fundamental step in exploratory data analysis (EDA). It helps identify data quality issues, such as systematic missingness, which can bias downstream modeling and inform decisions about imputation strategies or feature exclusion.

Answer

Encode all categorical features using label encoding for simplicity.

Answer

Include all categorical features with high cardinality as-is in the model.

Question 12

A data scientist is trying to read a CSV file from S3 bucket 'my-bucket' with key 'training/data.csv' using an IAM role with the attached policy shown in the exhibit. The read operation fails with an Access Denied error. What is the most likely cause?

Accepted Answer

The policy does not include the s3:ListBucket permission, which is required to access the object.. The s3:GetObject permission alone is insufficient to read an object from S3 when the request is made via the AWS Console or certain SDK operations that first list the bucket's contents. The s3:ListBucket permission is required for the ListObjects API call, which is often implicitly invoked to resolve the object key path. Without it, the read operation fails with an Access Denied error even if the GetObject permission is granted.

Answer

The object is encrypted with SSE-KMS and the role does not have kms:Decrypt permission.

Answer

The resource ARN in the first statement should be 'arn:aws:s3:::my-bucket/training' without the wildcard.

Answer

The policy explicitly denies s3:GetObject because of the second statement with the trailing slash.

Question 13

A data scientist is building a fraud detection model using a dataset of 500,000 credit card transactions. The dataset contains 20 features, including transaction amount, merchant category, time since last transaction, and customer age. The target variable 'is_fraud' has 0.1% positive examples. Initial EDA reveals that the transaction amount distribution is highly skewed with a long tail. Also, there are missing values in the 'customer_age' field (5% missing). The data scientist needs to prepare the data for training a binary classifier. Which combination of preprocessing steps should the data scientist apply to address these issues and improve model performance? (Select TWO.)

Accepted Answer

Apply log transformation to the transaction amount to reduce skewness.. Option C is correct because applying a log transformation to the highly skewed transaction amount reduces skewness and compresses the dynamic range, which helps many machine learning algorithms (especially those sensitive to feature scales like logistic regression or SVM) converge faster and perform better. This is a standard technique for handling right-skewed distributions without losing data.

Answer

Use SMOTE to generate synthetic samples of the minority class.

Answer

Apply standard scaling to all numerical features.

Answer

Drop the transaction amount feature because of its skewness.

Question 14

A company is using Amazon SageMaker to tune hyperparameters for a gradient boosting model. The objective is to minimize root mean squared error (RMSE). The data scientist wants to explore the hyperparameter space efficiently. Which THREE hyperparameter tuning strategies should the data scientist consider? (Choose 3.)

Accepted Answer

Bayesian optimization. Bayesian optimization is correct because it builds a probabilistic model of the objective function (RMSE) and uses an acquisition function to select the next hyperparameter combination to evaluate. This approach is sample-efficient, making it ideal for expensive-to-evaluate models like gradient boosting, as it balances exploration and exploitation to find optimal hyperparameters with fewer trials.

Answer

Grid search

Answer

Manual search

Question 15

A machine learning engineer is training a neural network on Amazon SageMaker using a custom Docker container. The training job fails with an error: 'CUDA out of memory.' The training instance is an ml.p3.2xlarge with 16 GB GPU memory. The model and data fit into memory when using batch size 32, but the engineer wants to maximize GPU utilization. Which approach should the engineer use to fix the out-of-memory error while maintaining efficient training?

Accepted Answer

Implement gradient accumulation with a larger effective batch size. Gradient accumulation allows the engineer to simulate a larger effective batch size by accumulating gradients over multiple forward/backward passes before performing an optimizer step. This keeps the per-step memory footprint low (avoiding CUDA out-of-memory) while maintaining training dynamics similar to a larger batch, thus maximizing GPU utilization without crashing.

Answer

Enable mixed precision training

Answer

Reduce batch size to 1

Answer

Use a CPU-only instance

Question 16

A data scientist runs a SageMaker training job that fails with the above error. The S3 bucket and object exist, and the IAM role has s3:GetObject permission. What is the MOST likely cause?

Accepted Answer

The S3 object was uploaded with incorrect checksum or is corrupted. The error indicates that SageMaker cannot read the S3 object, even though the bucket and object exist and the IAM role has s3:GetObject permission. The most likely cause is that the object was uploaded with an incorrect checksum or is corrupted, which prevents SageMaker from verifying the integrity of the data during the training job initialization. SageMaker uses ETag (MD5 checksum) validation when reading objects, and a mismatch triggers a failure.

Answer

The training instance does not have internet access

Answer

The S3 bucket has versioning enabled and the object version is not specified

Answer

The IAM role lacks kms:Decrypt permission for an encrypted S3 object

Question 17

A data scientist is building a multi-class classification model with 10 classes. The dataset has 100,000 samples. After training a random forest with 100 trees, the model achieves 85% accuracy on the test set. However, the data scientist notices that for one rare class (1% of data), recall is only 5%. Which technique is MOST likely to improve recall for the rare class without significantly reducing overall accuracy?

Accepted Answer

Apply SMOTE to oversample the rare class in the training data. SMOTE (Synthetic Minority Oversampling Technique) generates synthetic samples for the rare class by interpolating between existing minority instances, which directly addresses the class imbalance. This increases the model's exposure to the rare class during training, improving recall without discarding data or significantly altering the overall class distribution, thus preserving overall accuracy.

Answer

Increase the number of trees to 500

Answer

Use stratified sampling only for the test set

Answer

Reduce the decision threshold for the rare class to 0.1

Question 18

Which THREE techniques can help reduce overfitting in a neural network trained on a small dataset?

Accepted Answer

Apply L2 weight regularization. L2 weight regularization (also known as weight decay) penalizes large weights by adding a term to the loss function proportional to the sum of squared weights. This forces the network to learn simpler patterns and reduces sensitivity to noise in the training data, which is especially helpful when the dataset is small and prone to overfitting.

Answer

Increase the number of hidden layers

Answer

Train for more epochs

Question 19

A company is deploying a real-time fraud detection system using a gradient boosting model on AWS SageMaker. The model uses 200 features and is trained on 50 GB of data. The inference latency requirement is under 10 ms per request. During load testing, the endpoint shows average latency of 15 ms. Which change is MOST likely to reduce latency below 10 ms?

Accepted Answer

Reduce the number of features to the top 50 based on feature importance. Reducing the number of features from 200 to the top 50 directly decreases the amount of data each inference request must process, which lowers both feature engineering overhead and model evaluation time. For gradient boosting models on SageMaker, fewer features mean fewer decision tree splits to traverse per prediction, which can significantly reduce latency without requiring hardware changes. This is the most direct and cost-effective way to meet the 10 ms requirement.

Answer

Switch to a GPU-based instance type

Answer

Increase the number of trees in the model

Answer

Use a larger batch size for inference

Question 20

A company is using Amazon SageMaker to train a large language model. The training job is taking too long. The data scientist wants to reduce training time without sacrificing model accuracy. Which THREE strategies are MOST appropriate?

Accepted Answer

Use mixed precision training (float16). Mixed precision training (float16) reduces memory usage and accelerates computation by using half-precision floating-point numbers for most operations, while maintaining a single-precision copy of critical parameters to preserve accuracy. This directly reduces training time on compatible GPUs (e.g., NVIDIA V100, A100) without sacrificing model quality, as the loss scaling technique prevents underflow in gradients.

Answer

Switch from GPU instance to CPU instance

Answer

Increase the maximum sequence length